Join us at Supercomputing 2013 in Denver,
November 17-22.

There’s a lot going on, and we hope to see you there!

REFRESHMENTS AND CONVERSATION

[OpenMP Booth #4117]

  • Tuesday and Wednesday at 4:00pm until it’s gone:
    Hot Pretzels and a Keg of cold local microbrew – Come for a hot snack and cold beer!

IN-BOOTH PRESENTATIONS [OpenMP Booth #4117]

We are holding five presentations about the OpenMP API in our booth – booth #4117. Each presentation will be followed by a business-card drawing for an OpenMP fleece vest. Come in for a presentation, enter your business card for a chance to win a prize, and then stick around for conversation.

Presentation slides will be posted to the Resources page.

  • Tuesday, Nov 19th, 11:15 – 11:30am “Integrating OpenMP into Embedded Applications” – Markus Levy and Sunita Chandrasekaran
  • Tuesday, Nov 19th, 2:15 – 2:30pm “Introduction to OpenMP 4.0” – Yonghong Yan
  • Wednesday, Nov 20th, 11:15 – 11:30am OpenMP Tasking Explained” – Ruud van der Pas
  • Wednesday, Nov 20th, 2:15 – 2:30pm “OpenMP Accelerator for TI Devices” – Eric Stotzer
  • Thursday, Nov 21st, 11:15 – 11:30am “OpenMP and NVIDIA” – Jeff Larkin

TUTORIALS AND SESSIONS:

  • Sunday, Nov 17th, 8:30am – 5:00pm [Room 407] Tutorial: “Hybrid MPI & OpenMP Parallel Programming” – Rolf Rabenseifner, Georg Hager, Gabriele Jost.

This tutorial analyzes the strengths and weaknesses of several parallel programming models on clusters of SMP nodes. MPI-3.0 introduced a new shared memory programming interface, which can be combined with MPI message passing and remote memory access on the cluster interconnect. It can be used for direct neighbor accesses similar to OpenMP or for direct halo copies, and enables new hybrid programming models. These models are compared with various hybrid MPI+OpenMP approaches and pure MPI.. More information. Session details

  • Sunday, Nov 17th, 8:30am – 5:00pm [Room 402] Tutorial: “Hands-on Introduction to OpenMP3” – Tim Mattson, Mark Bull, Mike Pearce

In this tutorial, we will cover the core features of the OpenMP 3.1 standard. This will be a hands-on tutorial. We expect students to use their own laptops (with Windows, Linux, or OS/X). We will have access to systems with OpenMP (a remote SMP server), but the best option is for students to load an OpenMP compiler onto their laptops before the tutorial. Session details

  • Monday, Nov 18th, 8:30am – 5:00pm  [Room is 406] Tutorial: “Advanced OpenMP Tutorial: Performance and 4.0 Features” – Christian Terboven, Michael Klemm, Ruud van der Pas, Bronis R. de Supinski

While we quickly review the basics of OpenMP programming, we assume attendees understand basic parallelization concepts and will easily grasp those basics. Our Advanced OpenMP Programming tutorial explores the implications of possible OpenMP parallelization strategies. We focus on performance aspects, such as data and thread locality on NUMA architectures, false sharing, and exploitation of vector units. We discuss language features in-depth, and close with an overview of the new OpenMP 4.0 directives for attached compute accelerators. Session details

 

 

  • Tuesday, Nov 19th, 5:15 – 7:00pm
    Poster Reception:
    “Task Profiling through OpenMP Runtime API and Tool Support” – Ahmad R. Qawasmeh, Abid M. Malik, Deepak Eachempati, Barbara M. Chapman

The introduction of tasks in the OpenMP programming model brings a new level of parallelism. This also creates new challenges with respect to its applicability through an event-based performance profiling. The OpenMP Architecture Review Board (ARB) approved an interface specification known as the “OpenMP Runtime API (ORA) for Profiling” to enable performance tools to interact with OpenMP programs. Session details.

 

 

  • Tuesday, Nov 19th, 5:30 – 7:00pm [Rooms 301/302/303] Birds of a Feather: “OpenMP Goes Heterogeneous With OpenMP 4.0” – Michael Wong, Bronis de Supinski, Barbara Chapman

The OpenMP ARB has worked hard to add significant new features to the OpenMP specification. These include features for large core counts and for execution on platforms with heterogeneous cores and accelerators. The extensive set of new features that have been defined in the recently released OpenMP 4.0 will be discussed by key members of the Language Committee. Session details.

  • Thursday, Nov 15th, 2:00 – 2:30pm [Room 155-C] Exhibitor Forum: “Announcing OpenMP API Version 4.0” – Michael Wong

An overview of the new features in OpenMP  API version 4.0 – including support for accelerators, error handling, thread affinity, tasking extensions, support for Fortran 2003, and more.  Session details